对已运行的kafka集群添加副本

在已经运行的kafka集群中(业务主题已创建),如果一开始配置的默认副本是1,则后续再去修改配置文件为3,也只是针对后续新增的主题有效。原先主题不会生效。需要通过json文件去修改。

先创建__consumer_offsets主题的JSON

注意json文件中的"replicas":2,1,0这些是集群中的broke,有可能有的broke是1,2,3

新建一个consumer_offsets.json的文件,内容如下

复制代码
{
"version": 1,
"partitions": [
{"topic": "__consumer_offsets","partition": 0,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 1,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 2,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 3,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 4,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 5,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 6,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 7,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 8,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 9,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 10,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 11,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 12,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 13,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 14,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 15,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 16,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 17,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 18,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 19,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 20,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 21,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 22,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 23,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 24,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 25,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 26,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 27,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 28,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 29,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 30,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 31,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 32,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 33,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 34,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 35,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 36,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 37,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 38,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 38,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 39,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 40,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 41,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 42,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 43,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 44,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 45,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 46,"replicas": [1,0]},
{"topic": "__consumer_offsets","partition": 47,"replicas": [2,1]},
{"topic": "__consumer_offsets","partition": 48,"replicas": [0,2]},
{"topic": "__consumer_offsets","partition": 49,"replicas": [1,0]}
]
}

在任意一个节点上执行(只需在一个节点上执行一次)

bash 复制代码
bin/kafka-reassign-partitions.sh --zookeeper 192.168.1.1:2181 --reassignment-json-file consumer_offsets.json --execute

之后再加业务主题副本,新建一个doorinout.json的文件,内容如下

bash 复制代码
{
"version": 1,
"partitions": [
{"topic": "doorinout","partition": 0,"replicas": [2,1]},
{"topic": "doorinout","partition": 1,"replicas": [0,2]},
{"topic": "doorinout","partition": 2,"replicas": [1,0]},
]
}

在任意一个节点上执行(只需在一个节点上执行一次)

bash 复制代码
bin/kafka-reassign-partitions.sh --zookeeper 192.168.1.1:2181 --reassignment-json-file doorinout.json --execute
相关推荐
Thomas.Sir3 天前
第21课:PyTorch|GPU多卡训练与分布式训练基础【让多卡并行成为你的加速引擎】
人工智能·pytorch·分布式
Cicada1283 天前
库存消息消费的正确性设计——从幂等窗口到批量流水线
分布式·系统架构
Nano叶落3 天前
Kafka 消费积压排查入门:Docker 搭环境亲手制造一次 ‘消息堵死‘,10 分钟看懂 Lag
kafka
Francek Chen4 天前
【大数据处理与分析】数据仓库Hive:04 数据仓库Hive概述
大数据·数据仓库·hive·hadoop·分布式
Gl�ria4 天前
Hadoop/YARN 集群缩容:下线DN节点
大数据·hadoop·分布式
吉甫作诵4 天前
Kafka 集群安装与运维实战:消费组排查、Offset 重置与副本重分配
大数据·运维·分布式·kafka·消息队列
imDwAaY4 天前
消息队列四大核心问题:顺序性、幂等性、可靠性与一致性
学习·kafka·rabbitmq
wno7045 天前
Spring Boot整合Kafka
spring boot·kafka
是Dream呀5 天前
Harness 工程:让 Agent 真正把任务做完
人工智能·分布式·缓存·agent